230 ALASKAN GLACIER STUDIES
and ice streams flowed across divides now discharging glaciers in two directions. Where fiord water now stands great glaciers 2000 or 3000 feet deep flowed through the fiord valleys (PI. XCIE, A); and where even the great glaciers, like the Hubbard, now terminate, the ice rose 1000 to 2000 feet higher than at present. Even such great ice streams as the Hubbard, Turner and Nunatak Glaciers are mere dwindling glaciers compared to their predecessors on the same sites. It was a region of mighty glaciers, flowing irresistibly along mountain valley courses, eating away the valley walls and scouring the valley bottoms; and this condition extended along the Alaskan coast from British Columbia to the Alaska Peninsula.
After its long, but unknown period of advance, the glacier system began a period of recession whose detailed history is also unknown. There may have been minor advances and recessions, or it may have been one continuous period of recession. By it the glacier fronts were caused to recede even farther than at present, and the forest extended farther up the fiord than now, and even into the region where glaciers now stand. From this we infer that both the length of time of recession and the extent of recession were greater than the present period of recession. But by this it is not meant to intimate that the period of recession was a really long one, for little gorge erosion has been accomplished by the streams issuing from hanging valleys that were produced by the ice erosion of the preceding great advance. A couple of centuries, or perhaps twice that amount, would suffice to explain the phenomena observed, and much more than that would be too great a timo for the small work of stream erosion accomplished. Whether a similar period of recession was in progress in other portions of Alaska, or whether it was confined to the Yakutat Bay region and vicinity cannot now be stated.
During this period of recession, and perhaps mainly during the advance of the glaciers which terminated it, there was extensive deposition of gravels in Yakutat Bay, and especially in Russell Fiord. Then came a notable advance, pushing the glacier fronts once more far into Russell Fiord (PL XCIT, B) and Disenchantment Bay, but not nearly out to the point reached by the earlier great advance. It was of brief duration, and the work performed was not notable. The valley walls were freshly scoured, and probably there was effective glacial erosion in parts of the fiord bottoms; but the ice erosion did not suffice to remove the gravels deposited between the two advances, though it greatly sculptured them. This advance was a local phenomenon, not occurring in Prince William Sound to the northwest, nor in the Inside Passage, though a similar, and probably contemporaneous advance is recorded in the Muir Glacier region and north of the St. Elias Range. The advance in the Yakutat Bay region reached its maximum in a very recent period, for the vegetation growing on a lake beach formed by the advanced glacier at the head of Russell Fiord is not over a hah* century old.
Following this notable, though brief advance came the recession which was still in progress at the time of our studies in 1905 and 1906. The recession was so rapid that vegetation had not been able to follow it closely, the overridden gravels near the glaciers having only scattered plants, while even the outermost portions of the overridden gravels bore only a young growth of cottonwood and alders. It is certain that the glaciers began recession from their outermost stand not over a century ago and perhaps not much more than half a century ago. The greatest recession has occurred in the tidal glaciers, notably the Nunatak, and in the Hidden Glacier. The least recession has occurred in the piedmont glacier? and piedmont bulbs. In fact, it is possible that the piedmont con-